The Basics of a Content Delivery Network
Image shown: Huawei, Networking World, May 13, 2021
Content Delivery Network - What is it?
A Content Delivery Network (CDN) is a technology service used by Cloud Providers to provide low latency – faster response time – to their customers. Low latency is achieved by incorporating CDNs into architectural designs which route traffic along a private link instead of routing traffic along the internet.
The service basically consists of using caching to store copies of data on servers that are in closer proximity to customers than the origin servers housed in the main/regional data center. Copies of data are cached within the servers of CDNs, thereby providing low latency. Furthermore, using CDNs provide for Distributed Denial of Service (Ddos) protection and decreased CPU loads on origin servers. Using CDNs to provision for all of these benefits can result in tremendous cost savings for businesses.
Content Delivery Network – When to use It?
It is most beneficial to incorporate a CDN within architectural designs that require low latency to the extent that a millisecond difference in time is likely to have an adverse effect on business profitability. Companies that are dependent upon currency exchanges, stock exchanges, real time data analytics and other low latency driven transactions should utilize CDNs.
CDNs are most practical and cost effective for static content such as websites, distributions, and images. Static content is not updated frequently and using CDNs for this type of content ensures that the latest copy is available in the cache. In addition, there is significant cost savings by caching static content because the cost to update and configure the cache is minimized.
Content Delivery Network – How does it work?
In the case of a customer requesting a very popular image from a frequently used website, a CDN would work as follows: The first request made by any customer for the image, within a preset time period, would go through the CDN and be routed to the origin server. In response to the request, a copy of the image would be routed back to and cached within the CDN. Subsequent request for the image would go through the CDN and the copy would be routed to the customer because it is readily available in the cache. Once the preset time period has expired on the cache, an initial request for the image would be routed to the server of origin and the process would repeat itself.
Preset time periods of the cache are easily changed manually and in fact, there are times when the Cloud Provider or customer would need to cancel and reset the cache. An example of this would be when time sensitive data is updated. In this case, it’s crucial to have the latest information and canceling and resetting the cache would become pertinent.
Content Delivery Network – Why does it work?
Utilizing a guaranteed and consistent network gives traffic routing of CDNs high performance. Once a customer sends a request through a CDN, the request is now routed across a private link which results in the greatest advantage of using a CDN; low latency. The speeds of both the request and the response are maximized across the private network.
The performance of the internet pale in comparison. Rather than providing a guaranteed network for traffic, the internet provides for a “best effort” network for traffic. Traffic is frequently routed inefficiently across multiple hops and internet service providers. Therefore, the internet’s performance is less than ideal for low latency architectural designs.
Content Delivery Network – Are there additional benefits?
Besides providing low latency, CDNs are effective in preventing Ddos attacks from reaching origin servers housed within main/regional data centers. Preventing Ddos attacks can save businesses millions of dollars per attack. During a Ddos attack an exorbitant amount of request are sent to servers, the servers scale out, the servers become overloaded and eventually a portion(s) of the system or the entire system can slow down significantly or fail. Including a CDN within the architectural design prevents Ddos attacks because repeated and/or “bad” request are recognized locally and never routed to the origin server. The CDN serves as an alarm and protector, of sorts; and a customer can seamlessly be routed to other local servers within the CDN while the Ddos attack is being mitigated.
An additional benefit of using a CDN is that businesses can potential lower server usage cost charged by Cloud Providers. By caching copies of data within a CDN, request are responded to by the CDN, request to origin servers decrease and the CPU load of origin servers can be lowered. If the origin servers are responding to a reduced amount of request, they will not need to scale out as often and usage cost could be reduced.
Content Delivery Network – Need to access Cloud Provider specific service details?
Whitepapers – in depth descriptions of technology – expounding upon Content Delivery Networks are available for research by all Cloud Providers. Listed below are some popular providers with the name of their CDN service for easy reference.
Google – Google Cloud CDN
AWS – Amazon CloudFront
Azure – Microsoft Azure CDN
Oracle – Oracle Cloud Infrastructure (CDN)
Salesforce- Experience Cloud CDN
12/18/2021
Rev: 5/31/2022